Date: Mon, 23 Feb 1998 01:40:03 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: phk@critter.freebsd.dk (Poul-Henning Kamp) Cc: grog@lemis.com, jkh@time.cdrom.com, nate@mt.sri.com, tlambert@primenet.com, current@FreeBSD.ORG Subject: Re: More breakage in -current as a result of header frobbing. Message-ID: <199802230140.SAA16400@usr08.primenet.com> In-Reply-To: <12512.888186392@critter.freebsd.dk> from "Poul-Henning Kamp" at Feb 22, 98 11:26:32 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> >Back to the original subject, guys. I proposed the following > > > > [...] > > I don't think it will fly, I think trying to synchronize the factors > which control the time our hackers have to devote to FreeBSD would > be a major mistake... I agree. The unfortunate effect of following the suggestion, even if the time window control weren't a factor, would be to trigger all commits in a relatively short window. This is exactly the type of situation which will result in Bob and Fred committing conflicting changes that result in an unbuildable tree. Without some software-enforced procedural mechanism for avoiding this type of collision, hashing commits over "all available time" is as good a method as any. At least you get statistical protection, if nothing else. If you get a hash collision (ie: Bob and Fred's commits overlap), well, then you're potentially screwed, and if so, then it's going to require actual work to get the tree buildable. Not that this happening with an artificially bloated frequency wouldn't make my cvs locking proposal look a lot more desirable... ;-). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199802230140.SAA16400>